Search Results (7,932)

Background/Objectives: Multimodal large language models (LLMs) may directly interpret medical images, including thyroid ultrasounds (USs). Whether these models can reliably assess thyroid nodules—where subtle echogenic and morphological details are critical—remains uncertain. The American College of Radiology (ACR) TI-RADS system provides a structured framework for benchmarking artificial intelligence. This study evaluates GPT-5.0’s ability to interpret thyroid US images according to TI-RADS criteria and contextualizes its performance relative to expert radiologist assessment, using FNA cytology as the reference standard. Methods: This retrospective study included 100 patients (mean age 49.8 ± 12.6 years; 72 women) with cytology-confirmed diagnoses: Bethesda II (benign) or Bethesda V–VI (malignant). Each nodule had longitudinal and transverse US images acquired with high-frequency linear probes. A board-certified radiologist (>10 years’ experience) and GPT-5.0 independently assessed TI-RADS features (composition, echogenicity, shape, margin, echogenic foci) and assigned final categories. Agreement was analyzed using Cohen’s κ, and diagnostic performance was calculated using TR4–TR5 as positive for malignancy. Results: Agreement was substantial for composition (κ = 0.62), shape (κ = 0.70), and margin (κ = 0.68); moderate for echogenicity (κ = 0.48); and poor for echogenic foci (κ = 0.12). GPT-5.0 demonstrated a systematic, risk-averse tendency to up-classify nodules, leading to increased TR4–TR5 assignments. Overall, the TI-RADS agreement was 58% (κ = 0.31). The radiologist showed superior diagnostic performance (sensitivity 89%, specificity 85%) compared with GPT-5.0 (sensitivity 67%, specificity 49%), largely driven by false-positive TR4 classifications among benign nodules. Conclusions: GPT-5.0 recognizes several high-level TI-RADS features but struggles with microcalcifications and tends to overestimate malignancy risk within a risk-stratification framework, limiting its standalone clinical use. Ultrasound-specific training and domain adaptation may enable meaningful adjunctive roles in thyroid nodule assessment. Full article

Lipodystrophy is a rare group of metabolic disorders characterized by the abnormal distribution of body fat, which can lead to various metabolic complications due to the body’s inability to adequately process carbohydrates and fat. We report the case of a female, aged 53 years, who was admitted as an outpatient for progressive weight loss of the upper part of the body (face, neck, arms, and chest), dyspeptic complaints, fatigue, mild insomnia, and anxious behavior. Her medical history was characterized by the presence of dyslipidemia, hypertension, and a minor stroke episode. However, she denied any family-relevant medical history. Although the clinical perspective suggested a possible late onset of partial acquired lipodystrophy, due to the imaging exam that revealed an enlarged liver with inhomogeneous structure with multiple nodular lesions, scattered over both lobes, a lot of lab work-ups and complementary studies were performed. Eventually, a liver biopsy was performed by a laparoscopic approach during cholecystectomy, the histology consistent with metabolic disease-associated steatohepatitis (MASH). In conclusion, given their heterogeneity and rarity, lipodystrophies may be either overlooked or misdiagnosed for other entities. Barraquer–Simons syndrome (BSS) may be associated with liver disease, including cirrhosis and liver failure. Liver lipodystrophy in BSS may sometimes feature steatosis with a focal, multi-nodular aspect, multiplying the diagnostic burden. Liver lipodystrophy may manifest as asymptomatic fat accumulation but may progress to severe conditions, representing one of the major causes of mortality in BSS, apart from the cardio-vascular comorbidities. Given the potential of severe outcomes, it is mandatory to correctly assess the stage of liver disease since the first diagnosis. Full article

Explainability features are intended to provide insight into the internal mechanisms of an Artificial Intelligence (AI) device, but there is a lack of evaluation techniques for assessing the quality of provided explanations. We propose a framework to assess and report explainable AI features in medical images. Our evaluation framework for AI explainability is based on four criteria that relate to the particular needs in AI-enabled medical devices: (1) Consistency quantifies the variability of explanations to similar inputs; (2) plausibility estimates how close the explanation is to the ground truth; (3) fidelity assesses the alignment between the explanation and the model internal mechanisms; and (4) usefulness evaluates the impact on task performance of the explanation. Finally, we developed a scorecard for AI explainability methods in medical imaging that serves as a complete description and evaluation to accompany this type of device. We describe these four criteria and give examples on how they can be evaluated. As a case study, we use Ablation CAM and Eigen CAM to illustrate the evaluation of explanation heatmaps on the detection of breast lesions on synthetic mammographies. The first three criteria are evaluated for task-relevant scenarios. This framework establishes criteria through which the quality of explanations provided by medical devices can be quantified. Full article

In order to solve the dynamic analysis and interactive imaging control problems in the deformation process of bionic soft lenses, dielectric elastomer (DE) actuators are separated from a convex lens, and data-driven eye-controlled motion technology is investigated. According to the DE properties, which are consistent with the deformation characteristics of hydrogel electrodes, the motion and deformation effect of eye-controlled lenses under film prestretching, lens size, and driving voltage, is studied. The results show that when the driving voltage increases to 7.8 kV, the focal length of the lens, whose prestretching λ is 4, and the diameter d is 1 cm, varies in the range of 49.7 mm and 112.5 mm. And the maximum focal-length change could reach 58.9%. In the process of eye controlling design and experimental verification, a high DC voltage supply was programmed, and eye movement signals for controlling the lens were analyzed by MATLAB software (R2023b). Eye-controlled interactive real-time motion and tunable imaging of the lens were realized. The response efficiency of soft lenses could reach over 93%. The adaptive lens system developed in this research has the potential to be applied to medical rehabilitation, exploration, augmented reality (AR), and virtual reality (VR) in the future. Full article

Effective and transparent medical diagnosis relies on accurate and interpretable classification of medical images across multiple modalities. This paper introduces an explainable multi-modal image analysis framework based on a dual-stream architecture that fuses handcrafted descriptors with deep features extracted from a custom MobileNet. Handcrafted descriptors include frequency-domain and texture features, while deep features are summarized using 26 statistical metrics to enhance interpretability. In the fusion stage, complementary features are combined at both the feature and decision levels. Decision-level integration combines calibrated soft voting, weighted voting, and stacking ensembles with optimized classifiers, including decision trees, random forests, gradient boosting, and logistic regression. To further refine performance, a hybrid class-specific feature selection strategy is proposed, combining mutual information, recursive elimination, and random forest importance to select the most discriminative features for each class. This hybrid selection approach eliminates redundancy, improves computational efficiency, and ensures robust classification. Explainability is provided through Local Interpretable Model-Agnostic Explanations, which offer transparent details about the ensemble model’s predictions and link influential handcrafted features to clinically meaningful image characteristics. The framework is validated on three benchmark datasets, i.e., BTTypes (brain MRI), Ultrasound Breast Images, and ACRIMA Retinal Fundus Images, demonstrating generalizability across modalities (MRI, ultrasound, retinal fundus) and disease categories (brain tumor, breast cancer, glaucoma). Full article

Machine Learning (ML) and Deep Learning (DL) have shown significant promise in healthcare, particularly in medical image segmentation, which is crucial for accurate disease diagnosis and treatment planning. Despite their potential, challenges such as data privacy concerns, limited annotated data, and inadequate training data persist. Decentralized learning approaches such as federated learning (FL), split learning (SL), and split federated learning (SplitFed/SFL) address these issues effectively. This paper introduces “MedSegNet10,” a publicly accessible repository designed for medical image segmentation using split-federated learning. MedSegNet10 provides a collection of pre-trained neural network architectures optimized for various medical image types, including microscopic images of human blastocysts, dermatoscopic images of skin lesions, and endoscopic images of lesions, polyps, and ulcers. MedSegNet10 implements SplitFed versions of ten established segmentation architectures, enabling collaborative training without centralizing raw data and labels, reducing the computational load required at client sites. This repository supports researchers, practitioners, trainees, and data scientists, aiming to advance medical image segmentation while maintaining patient data privacy. Full article

The classification of breast cancer using machine learning techniques has become a critical tool in modern medical diagnostics. This study analyzes the performance of hybrid models that combine traditional machine learning algorithms (TMLAs) with a convolutional neural network (CNN)-based VGG16 model for feature extraction to improve accuracy for classifying eight breast cancer subtypes (BCS). The methodology consists of three steps. First, image preprocessing is performed on the BreakHis dataset at 400× magnification, which contains 1820 histopathological images classified into eight BCS. Second, the CNN VGG16 is modified to function as a feature extractor that converts images into representative vectors. These vectors constitute the training set for TMLAs, such as Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Naive Bayes (NB), leveraging VGG16’s ability to capture relevant features. Third, k-fold cross-validation is applied to evaluate the model’s performance by averaging the metrics obtained across all folds. The results reveal that hybrid models leveraging a CNN-based VGG16 model for feature extraction, followed by TMLAs, achieve accuracy outstanding experimental accuracy. The KNN-based hybrid model stood out with a precision of 0.97, accuracy of 0.96, sensitivity of 0.96, specificity of 0.99, F1-score of 0.96, and ROC-AUC of 0.97. These findings suggest that, with an appropriate methodology, hybrid models based on TMLA have strong potential in classification tasks, offering a balance between performance and predictive capability. Full article

Background: Sodium–glucose cotransporter 2 inhibitors (SGLT2 inhibitors), initially developed for glycemic control in type 2 diabetes, have demonstrated robust cardiovascular and renal benefits. Emerging evidence suggests that these agents may also affect valvular pathobiology, particularly in degenerative aortic stenosis (AS), through anti-inflammatory and antifibrotic mechanisms. Objectives: This study evaluated whether SGLT2 inhibitor use is associated with improved clinical outcomes in degenerative AS, including all-cause mortality and the need for SAVR or TAVR, recognizing that these endpoints represent surrogate rather than direct measures of valve hemodynamic progression. Methods: A retrospective cohort analysis was conducted using TriNetX, a federated electronic medical record-based research network. Diagnoses are captured using ICD-9/ICD-10-CM codes and medications using ATC codes. Adults with non-rheumatic AS were stratified by SGLT2 inhibitors use. Propensity score matching (1:1) was performed to balance baseline characteristics between treated and untreated groups (n = 10,912 per group). Primary outcomes included all-cause mortality, TAVR, and SAVR during follow-up. Echocardiographic parameters (AVA, Vmax, mean gradient) were not systematically available. Results: After adjustment for comorbidities, SGLT2 inhibitor use was independently associated with lower all-cause mortality (6.15% vs. 9.34% HR 0.595; 95% CI 0.552–0.641; p < 0.001), TAVR (2.81% vs. 2.89% HR 0.835; 95% CI 0.746–0.934; p = 0.002), SAVR (1.28% vs. 1.90% HR 0.514; 95% CI 0.442–0.599; p < 0.001), cardiac arrest (0.82% vs. 1.21% HR 0.71; 95% CI 0.582–0.867; p < 0.001), and end-stage kidney disease (0.40% vs. 1.0% HR 0.292; 95% CI 0.222–0.384; p < 0.001). Although these associations may suggest slower disease progression, interpretation is limited by the lack of systematic echocardiographic follow-up. Conclusions: In addition to their established benefits in heart failure and renal protection, SGLT2 inhibitors may have valve-preserving effects in degenerative AS. Because true hemodynamic progression could not be evaluated, these results should be viewed as associations with surrogate clinical endpoints. Prospective studies with standardized imaging are required to determine whether SGLT2 inhibition can directly alter the course of this currently untreatable disease Full article

Endometrial lesions are common in gynecology, exhibiting considerable clinical heterogeneity across different subtypes. Although ultrasound imaging is the preferred diagnostic modality due to its noninvasive, accessible, and cost-effective nature, its diagnostic performance remains highly operator-dependent, leading to subjectivity and inconsistent results. To address these limitations, this study proposes a hierarchical feature selection and optimization framework for endometrial lesions, aiming to enhance the objectivity and robustness of ultrasound-based diagnosis. Firstly, Kernel Principal Component Analysis (KPCA) is employed for nonlinear dimensionality reduction, retaining the top 1000 principal components. Secondly, an ensemble of three filter-based methods—information gain, chi-square test, and symmetrical uncertainty—is integrated to rank and fuse features, followed by thresholding with Maximum Scatter Difference Linear Discriminant Analysis (MSDLDA) for preliminary feature selection. Finally, the Whale Migration Algorithm (WMA) is applied to population-based feature optimization and classifier training under the constraints of a Support Vector Machine (SVM) and a macro-averaged F1 score. Experimental results demonstrate that the proposed closed-loop pipeline of “kernel reduction—filter fusion—threshold pruning—intelligent optimization—robust classification” effectively balances nonlinear structure preservation, feature redundancy control, and model generalization, providing an interpretable, reproducible, and efficient solution for intelligent diagnosis in small- to medium-scale medical imaging datasets. Full article

Ionic liquids (ILs) have emerged as multifunctional compounds with low volatility, high thermal stability, and tunable solvation capabilities, making them highly promising for biomedical applications. First explored in the late 1990s and early 2000s for enhancing the thermal stability of enzymes, antimicrobial agents, and controlled release systems, ILs have since gained significant attention in drug delivery, antimicrobial treatments, medical imaging, and biosensing. This review examines the diverse functions of ILs in contemporary therapeutics and diagnostics, highlighting their transformative capabilities in improving drug solubility, bioavailability, transdermal permeability, and pathogen inactivation. In drug delivery, ILs improve solubility of bioactive compounds, with several IL formulations achieving substantial solubility enhancements for poorly soluble drugs. Bio-ILs, in particular, show promise in enhancing drug delivery systems, such as improving transdermal permeability. ILs also exhibit significant antimicrobial and antiviral activity, offering new avenues for combating resistant pathogens. Despite their broad potential, challenges such as cytotoxicity, long-term metabolic effects, and the stability of ILs in physiological conditions persist. While much research has focused on their physicochemical properties, biological activity and in vivo studies are still underexplored. The future directions for ILs in biomedical applications include the development of bioengineered ILs and hybrid ILs, combining functional components like nanoparticles and polymers to create multifunctional materials. These ILs, derived from renewable resources, show great promise in personalized medicine and clinical applications. Further research is necessary to evaluate their pharmacokinetics, biodistribution, and long-term safety to fully realize their biomedical potential. This study emphasizes the potential of ILs to transform therapeutic and diagnostic technologies by highlighting present shortcomings and offering pathways for clinical translation, while also debating the need for continuous research to fully utilize their biomedical capabilities. Full article

Prolonged +Gz acceleration during high-performance flight exposes pilots to the risk of G-induced loss of consciousness (G-LOC), a dangerous condition that compromises operational safety. To enable early detection without intrusive sensors, we present a vision-based warning system that analyzes facial action units (AUs) as physiological indicators of impending G-LOC. Our approach combines computer vision with physiological modeling to capture subtle facial microexpressions associated with cerebral hypoxia using widely available RGB cameras. We propose a novel Graph Convolution-Augmented Vision Transformer (GC-ViT) network architecture that effectively captures dynamic AU variations in pilots under G-LOC conditions by integrating global context modeling with vision Transformer. The proposed framework integrates a vision–semantics collaborative Transformer for robust AU feature extraction, where EfficientNet-based spatiotemporal modeling is enhanced by Transformer attention mechanisms to maintain recognition accuracy under high-G stress. Building upon this, we develop a graph-based physiological model that dynamically tracks interactions between critical AUs during G-LOC progression by learning the characteristic patterns of AU co-activation during centrifugal training. Experimental validation on centrifuge training datasets demonstrates strong performance, achieving an AUC-ROC of 0.898 and an AP score of 0.96, confirming the system’s ability to reliably identify characteristic patterns of AU co-activation during G-LOC events. Overall, this contact-free system offers an interpretable solution for rapid G-LOC detection, or as a complementary enhancement to existing aeromedical monitoring technologies. The non-invasive design demonstrates significant potential for improving safety in aerospace physiology applications without requiring modifications to current cockpit or centrifuge setups. Full article

Computer vision and deep learning are essential in diverse fields such as autonomous driving, medical imaging, face recognition, and object detection. However, enhancing low-light remote sensing images remains challenging for both research and real-world applications. Low illumination degrades image quality due to sensor limitations and environmental factors, weakening visual fidelity and reducing performance in vision tasks. Common issues such as insufficient lighting, backlighting, and limited exposure create low contrast, heavy shadows, and poor visibility, particularly at night. We propose QWR-Dec-Net, a quaternion-based Retinex decomposition network tailored for low-light image enhancement. QWR-Dec-Net consists of two key modules: a decomposition module that separates illumination and reflectance, and a denoising module that fuses a quaternion holistic color representation with wavelet multi-frequency information. This structure jointly improves color constancy and noise suppression. Experiments on low-light remote sensing datasets (LSCIDMR and UCMerced) show that QWR-Dec-Net outperforms current methods in PSNR, SSIM, LPIPS, and classification accuracy. The model’s accurate illumination estimation and stable reflectance make it well-suited for remote sensing tasks such as object detection, video surveillance, precision agriculture, and autonomous navigation. Full article

Mitral valve orifice area is a key functional metric that depends on complex geometric features, motivating a systematic assessment of the relative influence of these parameters. In this study, the mitral valve geometry is parameterized using twelve geometric variables, and a global sensitivity analysis based on Sobol indices is performed to quantify their relative importance. Because global sensitivity analysis requires many simulations, a Gaussian Process regressor is developed to efficiently predict the orifice area from the geometric inputs. Structural simulations of the mitral valve are carried out in Abaqus, focusing exclusively on the valve mechanics. The predicted distribution of orifice areas obtained from the Gaussian Process shows strong agreement with the ground-truth simulation results, and similar agreement is observed when only the most influential geometric parameters are varied. The analysis identifies a subset of geometric parameters that dominantly govern the mitral valve orifice area and can be reliably extracted from medical imaging modalities such as echocardiography. These findings establish a direct link between echocardiographic measurements and physics-based simulations and provide a framework for patient-specific assessment of mitral valve mechanics, with potential applications in guiding interventional strategies such as MitraClip placement. Full article

Each year, around 23 million miscarriages occur worldwide, which have a substantial emotional impact on parents, and impose significant societal costs. While medical care accounts for most expenses, work productivity loss contributes significantly. Addressing underlying causes of miscarriage could improve parents’ mental health and potentially their economic impact. In most countries, investigations into miscarriage causes are only recommended after recurrent cases, focusing mainly on maternal factors. Fetal and placental tissue are rarely examined, as current guidelines do not advise routine genetic analyses of pregnancy tissue, because the impact of further clinical decision making and individual prognosis is unclear. However, this leaves over 90% of all miscarriage cases unexplained and highlights the need for alternative methods. We therefore conducted a narrative review on genetic analysis, autopsy, and imaging of products of conception (POC). Karyotyping, QF-PCR, SNP array, and aCGH were reviewed in different research settings, with QF-PCR being the most cost-effective, while obtaining the highest technical success rate. Karyotyping, historically being considered the gold standard for POC examination, was the least promising. Post-mortem imaging techniques including post-mortem ultrasound (PMUS), ultra-high-field magnetic resonance imaging (UHF-MRI), and microfocus computed tomography (micro-CT) show promising diagnostic capabilities in miscarriages, with micro-CT achieving the highest cost-effective performance. In conclusion, current guidelines do not recommend diagnostic testing for most cases, leaving the majority unexplained. Although genetic and imaging techniques show promising diagnostic potential, they should not yet be implemented in routine clinical care and require thorough evaluation within research settings—assessing not only diagnostic and psychosocial outcomes but also economic implications. Full article

Aortic stenosis (AS) is the most prevalent valvular heart disease in Western countries and it is especially associated with older age. With its progressive course, AS leads to ventricular hypertrophy, impaired diastolic and systolic function, and symptomatic deterioration. The natural history of AS is closely linked to the extent of myocardial and extracardiac damage in association with the patients comorbidities. Diagnosis relies primarily on transthoracic echocardiography, which assesses valve morphology, quantifies stenosis severity, and evaluates cardiac remodeling. However, discordant grading is frequent, necessitating advanced imaging to clarify the severity and the mechanism of the stenosis and stratify risk. Treatment is predominantly interventional, as no medical therapy is able to stop disease progression. Surgical aortic valve replacement (SAVR) and transcatheter aortic valve replacement (TAVR) are the two treatment options. Special clinical scenarios—such as cardiogenic shock or concomitant cardiac amyloidosis—pose additional diagnostic and therapeutic challenges and require individualized, multidisciplinary management. Overall, contemporary AS care increasingly integrates multimodality imaging, refined risk stratification, and tailored interventional strategies to optimize outcomes. Full article